The overall goal of this Program Project is to develop safe and highly efficacious intervention strategies for prevention of melanoma nd non- melanoma skin cancers (NMSC) and to develop basic science and clinical research approaches which will serve as models for the chemoprevention of a wide range of human epithelial cancers. In this Project, we propose to demonstrate that specific histopathologic and morphometric abnormalities, genetic alterations, and immunohistochemical surrogate endpoint biomarker (SEB) changes are associated with a multi-step progression from normal skin to NMSC or melanoma in human study participants and that these biomarkers can be modulated safely by novel topical chemoprevention agents. Approximately 60% of squamous cell carcinomas arise from pre- existing actinic keratoses (AKs) and /or under contiguous skin surfaces. AKs may represent a significant risk factor for melanoma as well. Additionally, case-control studies found that dysplastic nevi (DN) are the strongest risk factor for melanoma. In the first year of this grant, subjects with DN, SCC, AKs, and pre-clinical AKs will be recruited to document that specific histopathologic and morphometric abnormalities, genetic alterations and immunohistochemical biomarkers are associated with the progression from normal skin to AK to SCC and normal skin to DN to melanoma and that these biomarkers are reproducible for at least a three month time period. Researchers agree that the probability of successfully altering the natural history of any cancer increases by targeting an earlier, rather than a later, time point in carcinogenesis. Therefore, in years 2-5, subjects with preclinical AKs or DN will be recruited to conduct phase IIa randomized placebo-controlled cancer prevention clinical trials of topical agents including difluoromethylornithine (DFMO), epigallocatechin gallate (EGCG), 9-cis-retinoic acid perillyl alcohol and sodium salicylate. Agents selected must first be shown: 1) active in a mouse UVB carcinogenesis model and/or a transgenic mouse melanoma model; 2) locally and systematically non-toxic in murine models; and 3) able to penetrate full thickness skin from BALB/c mice and humans (i.e. face lift skin). During these clinical studies we will determine the predictive accuracy of ultraviolet and polarized photography with respect to identification of abnormal histopathologic and morphometric areas of forearm skin epidermis and, ultimately, examine the histopathologic/morphometric and molecular genetic pathogenesis of skin preneoplasia.